arrow
返回

ISPDiff: Interpretable Scale-Propelled Diffusion Model for Hyperspectral Image Super-Resolution

delete2024-01-01
delete18
PRE
AI
W
Wenqian Dong
S
Sen Liu
Q
Qu, Jiahui
X
Xiao, Song *
Y
Yunsong Li
DOI:10.1109/TGRS.2024.3407967delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Hyperspectral image (HSI) super-resolution (SR) employing the denoising diffusion probabilistic model (DDPM) holds significant promise with its remarkable performance. However, existing relevant works exhibit two limitations: 1) directly applying DDPM to fusion-based HSI-SR ignores the physical mechanism of HSI-SR and unique characteristics of HSI, resulting in less interpretability and 2) scale-invariant DDPM suffers from a time-consuming inference. To tackle these issues, we propose an interpretable scale-propelled diffusion (ISPDiff) model for HSI-SR, which combines the underlying principles of HSI-SR with DDPM for progressively unrolling reconstruction by learning its distribution at various scales, enhancing the transparency significantly and reducing the inference time prominently. Concretely, we destroy and downsample HSI into Gaussian noise in the forward process of ISPDiff. Then we design a unified scale-flexible model in the backward process to iteratively refine HSI in a coarse-to-fine manner through scale-matched reconstruction and cross-scale upsampling, which can be unfolded with optimization algorithms. These solved equations are one-to-one corresponding unrolled into two deep neural networks, called progressive perceptual model-driven scale-matched restoration network (P2MSRN) and cross-scale model-driven upsampling network (CMUN). Through end-to-end training, the proposed ISPDiff implements HSI-SR with a scale-propelled unrolling diffusion characterized by enhanced interpretability, stronger task orientation, and reduced time consumption. Systematic experiments have been conducted on three public datasets, demonstrating that ISPDiff outperforms state-of-the-art methods. The code is available at https://github.com/Jiahuiqu/ISPDiff.
Keyword:
Image reconstruction
Hyperspectral imaging
Spatial resolution
Task analysis
Mathematical models
Transformers
Superresolution
Deep unrolling network
diffusion model
hyperspectral image (HSI)
super-resolution (SR)

期刊

IEEE Transactions on Geoscience and Remote Sensing 封面图
IEEE Transactions on Geoscience and Remote Sensing
IF:
8.6
论文数:
2.1W
被引数:
10.7W

机构

X
Xidian University
学者数:
2.4W
论文数: 1.9W
被引数: 9.7K
引用论文

引用论文

Inhibition of Tumor Promotion by a Biomimetic Superoxide Dismutase
err1983-07-01
err0
PREAI
errThomas W. Kensler; Donna M. Bush; Walter J. Kozumbo
err分享
err收藏
FusionNet: An Unsupervised Convolutional Variational Network for Hyperspectral and Multispectral Image Fusion
err2020-01-01
err81
PREAI
errWang, Zhengjue; Chen, Bo; Lu, Ruiying; Zhang, Hao; Liu, Hongwei; Varshney, Pramod K.
err分享
err收藏
LRR-Net: An Interpretable Deep Unfolding Network for Hyperspectral Anomaly Detection
err2023-01-01
err198
PREAI
errLi, Chenyu; Zhang, Bing; Hong, Danfeng; Yao, Jing; Chanussot, Jocelyn
err分享
err收藏
Natural course of post COVID-19 condition and implications for trial design and outcome selection: A population-based longitudinal cohort study
err
IF0
err2022-06-22
err0
errOAAI
errTala Ballouz; Dominik Menges; Alexia Anagnostopoulos; Anja Domenghino; Hélène E Aschmann; Anja Frei; Jan S Fehr; Milo A Puhan
err分享
err收藏
学者 查看更多内容